CN103914842A - Foreground detecting method based on Codebook background differencing - Google Patents
Foreground detecting method based on Codebook background differencing Download PDFInfo
- Publication number
- CN103914842A CN103914842A CN201410136069.7A CN201410136069A CN103914842A CN 103914842 A CN103914842 A CN 103914842A CN 201410136069 A CN201410136069 A CN 201410136069A CN 103914842 A CN103914842 A CN 103914842A
- Authority
- CN
- China
- Prior art keywords
- box
- value
- background
- codebook
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a foreground detecting method based on Codebook background differencing. The method comprises the following steps: conducting pre-processing on a video sequence and reinforcing the edge of a target; conducting background modeling on the video sequence after pre-processing by utilizing Codebook background differencing; conducting background updating on the video sequence by utilizing the Codebook background differencing; conducting background detection on a current frame and a background frame through the Codebook background differencing to acquiring a moving target in a current scene. By means of the foreground detecting method based on the Codebook background differencing, the foreground detection of foreign matter invasion into a high-tension transmission line is achieved. Compared with a traditional target detecting algorithm, the foreground detecting method based on the Codebook background differencing has higher accuracy and robustness.
Description
Technical field
The present invention relates to a kind of foreground detection method based on Codebook background subtraction point-score, particularly relate to a kind of foreground detection method based on Codebook background subtraction point-score to ultra-high-tension power transmission line foreign body intrusion.
Background technology
Ultra-high-tension power transmission line is the critical facility of electric system, and as the core component of electrical network, its safety and reliability concerns the stable of whole electric system.Intelligent grid requires on-line monitoring system to introduce Video Supervision Technique, realizes the target of ultra-high-tension power transmission line Real-Time Monitoring.It is larger that traditional artificial line walking mode is affected by patrol officer's subjective factor, easily occurs the situations such as undetected and flase drop, is difficult to guarantee the accuracy of testing result.A large amount of manpower and materials can be expended simultaneously, the requirement of real time on-line monitoring can not be met.Can not get rid of timely and effectively the potential safety hazard of transmission line of electricity and shaft tower, cause transmission line of electricity and shaft tower often to have fault and occur, serious power grid security accidents such as transmission line of electricity broken string, shaft tower inclination, cause great national economy loss.
Under dynamic system, the general frame of video monitoring process generally comprises following step: system environments modeling, moving object detection, target classification, target following, goal behavior are understood and multi-video camera information merges.Such as, and the object of target detection is in image sequence, surveyed area corresponding to moving target foreign matters such as () cranes to be split.Extracting after foreground target (moving target), only notice need to be placed in the processing of foreground area, and can not go to consider that other do not comprise the region of useful information.But MOG(gauss hybrid models) and nonparametric model due to the restriction of internal memory, can not be applied in and need for a long time under bulk sampling background situation, such as blowing hard in jungle, the scene etc. of snowing heavily, is not suitable for the foreground detection of ultra-high-tension power transmission line.
Summary of the invention
The deficiency existing for overcoming above-mentioned prior art, one of the present invention object is to provide a kind of foreground detection method based on Codebook background subtraction point-score, it is by setting up a code book for each pixel, the statistical information of statistical pixel values, by the statistics of the historical region of pixel value being set up to the distributed areas of pixel value, and distinguish pixel value with the value of present frame respective pixel with respect to the position of distributed areas and belong to foreground target or background, realize the foreground detection to ultra-high-tension power transmission line foreign body intrusion, there is higher degree of accuracy and robustness than traditional algorithm of target detection.
For reaching above-mentioned and other object, the present invention proposes a kind of foreground detection method based on Codebook background subtraction point-score, comprises the steps:
Step 1, carries out pre-service to video sequence, and object edge is strengthened;
Step 2, utilizes Codebook method to carry out background modeling to pretreated video sequence;
Step 3, utilizes Codebook method to carry out context update to this video sequence;
Step 4, carries out foreground detection by present frame and background frames by Codebook background subtraction point-score, obtains the moving target in current scene.
Further, step 2 further comprises the steps:
Step 2.1, sets up a code book to the each pixel in video sequence, and this code book is made up of some boxes;
Step 2.2, each box on each axle defines by two groups of threshold values, and these two groups of threshold values are boundary threshold and study threshold value, and each group threshold value forms by two elements, is respectively maximal value and minimum value;
Step 2.3, if pixel (x, y) is at the value B of new background model B
(x, y)fall the study threshold value of i box of pixel (x, y)
between, the boundary threshold of i box
to expand to comprise the value B into new background model
(x, y), and study threshold value
according to new value B
(x, y)border (B
(x, y)-g, B
(x, y)+ g) with study threshold value
position relationship adjust;
Step 2.4, if new background sample B
(x, y)outside the study threshold value of N box of pixel (x, y), will start to produce a new box, N+1 box of pixel (x, y) stores new value.
Further, in step 2.3, work as B
(x, y)-g is less than
time,
subtract one, otherwise constant, work as B
(x, y)+ g is greater than
time,
add one, otherwise also constant, wherein g is study range threshold, represents the study scope of box.
Further, the maximum boundary threshold value of newly-built box
with minimum boundary threshold
equate, be equal to current background model value B
(x, y), and maximum study threshold value
with minimum study threshold value
be initialized as respectively B
(x, y)+ g and B
(x, y)-g, the study scope that wherein g is box.
Further, the scope of study threshold value is greater than the scope of boundary threshold, and the study threshold value of box comprises boundary threshold.
Further, step 3 comprises the foundation of box and two steps of deletion of outmoded box, and the foundation of this box is identical with the process of setting up new box in step 2.
Further, the step of the deletion of this outmoded box is take the passive time as foundation, and each box has a state variable T
leave, enter this box distance now for recording the last new measured value, be called the passive time, exceed when the passive time threshold value T setting
lasttime, this box is deleted from background model.
Further, threshold value T
lastthe half of value learning time or frame number while being background modeling.
Further, in step 4, by the pixel value F of (x, y) position in present frame
(x, y)carrying out scope with N boxes of the pixel of same position in background model compares operation and belongs to foreground target or background to distinguish current pixel.
Further, by each pixel in present frame is carried out to identical operation, can obtain a two-value foreground image mask F
mask, Regional Representative's foreground target region that wherein pixel value is 255, the Regional Representative background area that pixel value is 0.
Compared with prior art, a kind of foreground detection method based on Codebook background subtraction point-score of the present invention is by setting up a code book for each pixel, the statistical information of statistical pixel values, by the statistics of the historical region of pixel value being set up to the distributed areas of pixel value, and distinguish pixel value with the value of present frame respective pixel with respect to the position of distributed areas and belong to foreground target or background, realize the foreground detection to ultra-high-tension power transmission line foreign body intrusion, there is higher degree of accuracy and robustness than traditional algorithm of target detection.
Accompanying drawing explanation
Fig. 1 is the flow chart of steps of a kind of foreground detection method based on Codebook background subtraction point-score of the present invention;
Fig. 2 is the overall flow figure of preferred embodiment of the present invention.
Embodiment
Below, by specific instantiation accompanying drawings embodiments of the present invention, those skilled in the art can understand other advantage of the present invention and effect easily by content disclosed in the present specification.The present invention also can be implemented or be applied by other different instantiation, and the every details in this instructions also can be based on different viewpoints and application, carries out various modifications and change not deviating under spirit of the present invention.
Fig. 1 is the flow chart of steps of a kind of foreground detection method based on Codebook background subtraction point-score of the present invention.As shown in Figure 1, a kind of foreground detection method based on Codebook background subtraction point-score of the present invention, comprises the steps:
Step 101, carries out pre-service to video sequence, except garbages such as denoisings, and object edge is strengthened.
Step 102, utilizes Codebook method to carry out background modeling to pretreated video sequence.
Specifically, step 102 comprises the steps:
Step 2.1, sets up a code book (Codebook) to the each pixel in video sequence, comprises one or a group code unit.Code book Codebook is made up of some boxes, and boxes can be understood as a dynamic interval range of pixel value, and these boxes comprise constant for a long time pixel value.It should be noted that, in preferred embodiment of the present invention, Codebook method is that each pixel is sampled, and arrives codebook set according to color distortion yardstick and brightness border cluster, and not all pixel has the code book quantity of equal number.Cluster representing by code book does not need corresponding single Gauss or other parameter distribution (even being single Gaussian distribution in the distribution of certain pixel), but utilizes some code books of corresponding that pixel.
Step 2.2, two groups of threshold values for each box on each axle (boundary threshold and study threshold value) define.Each group threshold value forms by two elements, is respectively maximal value and minimum value, and box is defined by following four threshold values, maximum boundary threshold value T
box_h, minimum boundary threshold T
box_l, maximum study threshold value T
learn_h, minimum study threshold value T
learn_l, the scope of its learning threshold value is greater than the scope of boundary threshold, and the study threshold value of box comprises boundary threshold.
Step 2.3, if pixel (x, y) is at the value B of new background model B
(x, y)fall pixel (x, y) i (i=0,1 ..., N) and the study threshold value of box
between, the boundary threshold of i box
to expand to comprise the value B into new background model
(x, y), and study threshold value
according to new value B
(x, y)border (B
(x, y)-g, B
(x, y)+ g) with study threshold value
position relationship adjust, work as B
(x, y)-g is less than
time,
subtract one, otherwise constant, work as B
(x, y)+ g is greater than
time,
add one, otherwise also constant, wherein g is study range threshold, represents the study scope of box, and empirical value is 10.
Step 2.4, if new background sample B
(x, y)outside the study threshold value of N box of pixel (x, y), will start to produce a new box, N+1 box of pixel (x, y) stores new value.The maximum boundary threshold value of newly-built box
with minimum boundary threshold
equate, be equal to current background model value B
(x, y), and maximum study threshold value
with minimum study threshold value
be initialized as respectively B
(x, y)+ g and B
(x, y)-g, the study scope that wherein g is box, adopts 10 equally.
Step 103, utilizes Codebook method to carry out context update to video sequence, establishes the background model of real-time update.Context update mainly comprises the foundation of following two step: box and the deletion of outmoded box.The process that the process that increases new box in CodeBook context update is set up new box during with background modeling is identical, no longer repeats here.The deletion of outmoded box, take the passive time as foundation, each box has a state variable T
leave, enter this box distance (or frame number) now for recording the last new measured value, be called the passive time.Exceed when the passive time threshold value T setting
lasttime, this box is deleted from background model, set T here
lastthe half of value learning time (or frame number) while being background modeling.
Step 104, carries out foreground detection by present frame and background frames by Codebook background subtraction point-score, just can obtain the moving target in current scene.Establishing after the background model of real-time update, by using present frame and background frames to carry out " it is poor to do ", just can obtain the moving target in current scene.In CodeBook background subtraction point-score, doing of the method for its " it is poor to do " and intermediate value and averaging method is poor different.In median method (or averaging method), it does poor operation is frame by frame pixel value to be carried out to reducing, as a result of pixel value of the absolute value of the difference of grey scale pixel value, and in CodeBook background subtraction point-score, the pixel value F of (x, y) position in present frame
(x, y), carry out scope with N boxes of the pixel of same position in background model and compare operation.
By contrast currency F
(x, y)the whether boundary threshold of some box in background model
in scope, distinguish current pixel and belong to foreground target or background.By each pixel in present frame is carried out to identical operation, can obtain a two-value foreground image mask F
mask, Regional Representative's foreground target region that wherein pixel value is 255, the Regional Representative background area that pixel value is 0.
Below will further illustrate the present invention by a specific embodiment: foreign body intrusion under ultra-high-tension power transmission line (as crane etc.) can cause danger to ultra-high-tension power transmission line.Affect at present near the multiple outside destroy hidden danger that one of safe and stable principal element of transmission line of electricity (electrical network) exists exactly transmission line of electricity; wherein mainly comprise the foreign body intrusion in the warning region of route protection; the architecture against regulations, fetch earth and quarry; the barbarous construction of crane, disaster and trees disaster etc.; overcoming outside destroy is current line security and stable key factor, the groundwork that line inspection is safeguarded especially.
Therefore, in order to detect real-time and accurately foreign matter, the present embodiment mainly adopts following scheme: first, utilize CodeBook method to form a Codebook(code book) to describe the fluctuating of pixel in background, observed reading current pixel and previous observed reading are compared, if two values are very approaching, it to be just judged as be the disturbance under the represented that color of previous observed reading, if two values are kept off, produce one group of new code book relevant to this pixel.Result can be envisioned as a branch of spot that swims in color space, and each spot represents the scope of a pixel value.The background of the present invention's structure can overcome the background object of disturbance, and can be good at detecting the target of slow motion.
Fig. 2 is the overall flow figure of preferred embodiment of the present invention.It comprises CodeBook background modeling, context update and CodeBook foreground detection technology, wherein, whether success is take background study threshold value as foundation in background initialization, in the time setting up the quantity of the successive image frame that background uses and exceed background study threshold value, judge background initialization success, select the required study frame number of background modeling according to the situation of scene, experience frame number is 50-300 frame; The basis for estimation of context update is context update threshold value, adopt refresh counter be carved into when from last context update current time the frame number (time) of process carry out record, in the time that the record of counter exceedes renewal threshold value, replace old model with new model, realize context update, upgrade threshold value and be made as 3-5 second.
As shown in Figure 2, the flow process of the present embodiment is:
(1) obtain video sequence, go to (2) or directly go to (7);
(2) carry out Codebook background modeling, go to (3);
(3) judge whether initialization success, if success goes to (4), otherwise goes to (2);
(4) determine whether context update, if desired context update, goes to (5), otherwise goes to (6);
(5) Codebook context update, goes to (6);
(6) set up background model, go to (7);
(7) Codebook foreground extraction, goes to (8);
(8) judge target prospect.
In sum, a kind of foreground detection method based on Codebook background subtraction point-score of the present invention is by setting up a code book for each pixel, the statistical information of statistical pixel values, by the statistics of the historical region of pixel value being set up to the distributed areas of pixel value, and distinguish pixel value with the value of present frame respective pixel with respect to the position of distributed areas and belong to foreground target or background, realize the foreground detection to ultra-high-tension power transmission line foreign body intrusion, there is higher degree of accuracy and robustness than traditional algorithm of target detection.
Tool of the present invention has the following advantages:
(1) not affecting under the prerequisite of segmentation effect, background model is carried out to high compression, thus the calculated amount of greatly reducing and storage space; Multiple experimental datas show, are that the life outdoor videos fragment that a section of 5 minutes, speed are 30fps is set up code book model, and average each pixel only needs 6.5 code words;
(2) because the variation of light is embodied in brightness conventionally, therefore adopt the method for brightness and colourity separate computations, simply and effectively solved local or overall light and changed the impact that background is produced;
(3) the code book structure stage allows the existence of prospect, but can be to prospect modeling; The background segment stage introduces buffering code book and upgrades, and adaptivity is better.
Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not for limiting the present invention.Any those skilled in the art all can, under spirit of the present invention and category, modify and change above-described embodiment.Therefore, the scope of the present invention, should be as listed in claims.
Claims (10)
1. the foreground detection method based on Codebook background subtraction point-score, comprises the steps:
Step 1, carries out pre-service to video sequence, and object edge is strengthened;
Step 2, utilizes Codebook method to carry out background modeling to pretreated video sequence;
Step 3, utilizes Codebook method to carry out context update to this video sequence;
Step 4, carries out foreground detection by present frame and background frames by Codebook background subtraction point-score, obtains the moving target in current scene.
2. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 1, is characterized in that, step 2 further comprises the steps:
Step 2.1, sets up a code book to the each pixel in video sequence, and this code book is made up of some boxes;
Step 2.2, each box on each axle defines by two groups of threshold values, and these two groups of threshold values are boundary threshold and study threshold value, and each group threshold value forms by two elements, is respectively maximal value and minimum value;
Step 2.3, if pixel (x, y) is at the value B of new background model B
(x, y)fall the study threshold value of i box of pixel (x, y)
between, the boundary threshold of i box
to expand to comprise the value B into new background model
(x, y), and study threshold value
according to new value B
(x, y)border (B
(x, y)-g, B
(x, y)+ g) with study threshold value
position relationship adjust;
Step 2.4, if new background sample B
(x, y)outside the study threshold value of N box of pixel (x, y), will start to produce a new box, N+1 box of pixel (x, y) stores new value.
3. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 2, is characterized in that: in step 2.3, work as B
(x, y)-g is less than
time,
subtract one, otherwise constant, work as B
(x, y)+ g is greater than
time,
add one, otherwise also constant, wherein g is study range threshold, represents the study scope of box.
4. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 3, is characterized in that: the maximum boundary threshold value of newly-built box
with minimum boundary threshold
equate, be equal to current background model value B
(x, y), and maximum study threshold value
with minimum study threshold value
be initialized as respectively B
(x, y)+ g and B
(x, y)-g, the study scope that wherein g is box.
5. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 2, is characterized in that: the scope of study threshold value is greater than the scope of boundary threshold, and the study threshold value of box comprises boundary threshold.
6. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 2, is characterized in that: step 3 comprises the foundation of box and two steps of deletion of outmoded box, and the foundation of this box is identical with the process of setting up new box in step 2.
7. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 6, is characterized in that: the step of the deletion of this outmoded box is take the passive time as foundation, and each box has a state variable T
leave, enter this box distance now for recording the last new measured value, be called the passive time, exceed when the passive time threshold value T setting
lasttime, this box is deleted from background model.
8. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 7, is characterized in that: threshold value T
lastthe half of value learning time or frame number while being background modeling.
9. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 2, is characterized in that: in step 4, by the pixel value F of (x, y) position in present frame
(x,
y)carrying out scope with N boxes of the pixel of same position in background model compares operation and belongs to foreground target or background to distinguish current pixel.
10. a kind of foreground detection method based on Codebook background subtraction point-score as claimed in claim 9, is characterized in that: by each pixel in present frame is carried out to identical operation, can obtain a two-value foreground image mask F
mask, Regional Representative's foreground target region that wherein pixel value is 255, the Regional Representative background area that pixel value is 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410136069.7A CN103914842A (en) | 2014-04-04 | 2014-04-04 | Foreground detecting method based on Codebook background differencing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410136069.7A CN103914842A (en) | 2014-04-04 | 2014-04-04 | Foreground detecting method based on Codebook background differencing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103914842A true CN103914842A (en) | 2014-07-09 |
Family
ID=51040498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410136069.7A Pending CN103914842A (en) | 2014-04-04 | 2014-04-04 | Foreground detecting method based on Codebook background differencing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103914842A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107944499A (en) * | 2017-12-10 | 2018-04-20 | 上海童慧科技股份有限公司 | A kind of background detection method modeled at the same time for prospect background |
| CN110422204A (en) * | 2019-07-23 | 2019-11-08 | 交控科技股份有限公司 | A kind of Train Dynamic time based on video analysis stops method and device |
| CN111126248A (en) * | 2019-12-20 | 2020-05-08 | 湖南千视通信息科技有限公司 | Method and device for identifying shielded vehicle |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622576A (en) * | 2011-01-31 | 2012-08-01 | 索尼公司 | Method and apparatus for background modeling, and method and apparatus for detecting background in video |
| CN103489196A (en) * | 2013-10-16 | 2014-01-01 | 北京航空航天大学 | Moving object detection method based on codebook background modeling |
-
2014
- 2014-04-04 CN CN201410136069.7A patent/CN103914842A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622576A (en) * | 2011-01-31 | 2012-08-01 | 索尼公司 | Method and apparatus for background modeling, and method and apparatus for detecting background in video |
| CN103489196A (en) * | 2013-10-16 | 2014-01-01 | 北京航空航天大学 | Moving object detection method based on codebook background modeling |
Non-Patent Citations (1)
| Title |
|---|
| 杨梦婕: "监控视频中的车辆检测与跟踪技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107944499A (en) * | 2017-12-10 | 2018-04-20 | 上海童慧科技股份有限公司 | A kind of background detection method modeled at the same time for prospect background |
| CN110422204A (en) * | 2019-07-23 | 2019-11-08 | 交控科技股份有限公司 | A kind of Train Dynamic time based on video analysis stops method and device |
| CN111126248A (en) * | 2019-12-20 | 2020-05-08 | 湖南千视通信息科技有限公司 | Method and device for identifying shielded vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103069434B (en) | For the method and system of multi-mode video case index | |
| CN104766086B (en) | The monitoring and managing method and system of a kind of way mark | |
| CN103491351A (en) | Intelligent video monitoring method for illegal buildings | |
| CN103150903B (en) | Video vehicle detection method for adaptive learning | |
| CN104463904A (en) | High-voltage line foreign matter invasion target detection method | |
| CN107679495A (en) | A kind of detection method of transmission line of electricity periphery activity engineering truck | |
| CN103578119A (en) | Target detection method in Codebook dynamic scene based on superpixels | |
| CN104268588B (en) | Railway wagon brake shoe pricker loses the automatic testing method of failure | |
| CN103093192A (en) | High voltage transmission line galloping identification method | |
| CN102750712A (en) | Moving object segmenting method based on local space-time manifold learning | |
| CN115761537B (en) | Power transmission line foreign matter intrusion identification method oriented to dynamic feature supplementing mechanism | |
| CN113255605A (en) | Pavement disease detection method and device, terminal equipment and storage medium | |
| CN103914842A (en) | Foreground detecting method based on Codebook background differencing | |
| CN103888731A (en) | Structured description device and system for mixed video monitoring by means of gun-type camera and dome camera | |
| CN103489012A (en) | Crowd density detecting method and system based on support vector machine | |
| Yuan et al. | Identification method of typical defects in transmission lines based on YOLOv5 object detection algorithm | |
| CN118038153A (en) | Method, device, equipment and medium for identifying external damage prevention of distribution overhead line | |
| CN105741277A (en) | ViBe (Visual Background Extractor) algorithm and SLIC (Simple Linear Iterative Cluster) superpixel based background difference method | |
| CN105741479A (en) | Integrated forest fire prevention IA-PCNN algorithm based on thermal imaging and smoke identification | |
| Chen et al. | Moving objects detection based on background subtraction combined with consecutive frames subtraction | |
| CN104574340A (en) | Video intrusion detection method based on historical images | |
| CN116740495A (en) | Training method and defect detection method for defect detection model of road and bridge tunnel | |
| CN113362330B (en) | Pantograph cavel real-time detection method, device, computer equipment and storage medium | |
| Zhou et al. | Intelligent identification method for natural disasters along transmission lines based on inter-frame difference and regional convolution neural network | |
| Jianfeng et al. | IMPLEMENTATION OF THE DUAL-BODY INTELLIGENT INSPECTION ROBOT IN SUBSTATION BASED ON DATA MINING ALGORITHM. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140709 |